Hello,
I am currently working on a project with a colleague using a Brain-Computer Music Interface (BCMI) to generate music from EEG signals. Affective states will be recorded with EEG signals and sent to a generative music algorithm. A couple of questions I had about the design:
1). We plan to use Emotiv Epoc+ (14 channels), does anyone know any way to run raw EEG data on EEGLAB in real-time, or must EEG data be recorded with external data acquisition software, and analyzed separately?
2). To run the generative music algorithm, is it necessary to train a classifier to model emotion? Can a SVM or random-forest classifier be used to classify emotion from EEG signals, which can then be fed into the generative music algorithm? Or is this step unnecessary?
3). We plan to use the DEAP dataset for emotion calibration. However, one difference my colleague and I had was whether:
- Pre-recorded EEG data from the DEAP dataset can be sent directly into the generative music algorithm? OR
- Is it necessary for participant's EEG data to be recorded, while being instructed to view items from the DEAP database to gauge their affective brain states?
DEAP dataset
http://www.eecs.qmul.ac.uk/mmv/datasets/deap/readme.html
The generative algorithm is inspired by Ehrlich et al. (2019), and is designed to generate sounds reflective of the user's affective state.
Article A closed-loop, music-based brain-computer interface for emot...
Hei! I am currently working on my ethics of BCI masters thesis. Would love to hear some more about your work.
All the best,
Ivana
@Pedro Morais Thank you. I had considered distributing my questions and based on the response turnout I will do that.
Hi I have been playing with BCI for music (using a Neurosky Mindwave headset until we can get the Emotiv one) and used BrainwaveOSC (it is open source on GitHub) to turn the EEG data into OSC. I found a MaxMSP patch to turn the OSC into MIDI, but not yet got it working. Would love to hear more about your work!
Caroline Stedman Interestingly, in our earlier prototype we used MaxMSP with MIDI, and have used MindYourOSC to calibrate the individual user's brain-state using push/pull command. We were able to produce musical sequences with MIDI patches, and had attempted to work with different scales corresponding with different emotions. However, this project came to a halt due to issues with the Epoc+, and we decided to expand with a more robust generative algorithm.
With Neurosky, it seems to be a similar design. All that I believe is needed is OSC ports to connect it to the patch on MaxMSP. However, I would recommend if you could hold off to get the Emotiv Epoc+ it would be best, due to its sampling rate and electrode placement. I'd be interested in learning more about your work.
- Badges
- Science topic
- Similar topics
- Psychology
- Emotion